Author: Logan Hall

Authors: Zijian Dong、Yilei Wu、Zijiao Chen、Yichi Zhang、Yueming Jin、Juan Helen Zhou Paper: https://arxiv.org/abs/2408.10567 Introduction In the realm of neuroimaging, the advent of large-scale, self-supervised pre-trained models for functional magnetic resonance imaging (fMRI) has shown significant promise in enhancing performance across various downstream tasks. However, fine-tuning these models is computationally intensive and time-consuming, often leading to overfitting due to the scarcity of training data. This study introduces Scaffold Prompt Tuning (ScaPT), a novel framework designed to efficiently adapt large-scale fMRI pre-trained models to downstream tasks with minimal parameter updates and improved performance. Related Work Fine-Tuning in fMRI Models Fine-tuning involves updating all pre-trained…

Authors: Sunil Arora、Sahil Arora、John D. Hastings Paper: https://arxiv.org/abs/2408.10351 Introduction In the digital age, social media has become an integral part of our daily lives, especially for teenagers. While these platforms are designed to enhance social interactions and improve our social lives by facilitating the sharing of personal experiences and life events, they often result in adverse psychological impacts. This study, conducted by Sunil Arora, Sahil Arora, and John D. Hastings, investigates the paradoxical outcomes of social media use among teenagers. The algorithms and structures that underlie social media platforms inadvertently contribute to profound psychological impacts, influencing teenagers in unforeseen ways.…

Authors: Jian Guo、Heung-Yeung Shum Paper: https://arxiv.org/abs/2408.10255 Building AGI for Quantitative Investment: The Large Investment Model (LIM) Introduction Quantitative investment, often referred to as “quant,” involves the use of mathematical, statistical, or machine learning models to drive financial investment strategies. These models execute trading instructions at speeds and frequencies unattainable by human traders. However, traditional quantitative investment research is facing diminishing returns due to rising labor and time costs. To address these challenges, the Large Investment Model (LIM) has been introduced as a novel research paradigm designed to enhance both performance and efficiency at scale. LIM employs end-to-end learning and universal…

Authors: Marianela Morales、Alberto Pozanco、Giuseppe Canonaco、Sriram Gopalakrishnan、Daniel Borrajo、Manuela Veloso Paper: https://arxiv.org/abs/2408.10889 Learning Action Costs from Input Plans: A Comprehensive Analysis Introduction In the realm of classical planning, the primary objective is to devise a sequence of deterministic actions that transition an initial state to a goal state. Traditionally, the focus has been on learning the dynamics of actions—how actions change the state—while assuming that the cost associated with each action is known. However, in many real-world applications, the cost of actions is either unknown or only approximately known. This paper introduces a novel problem: learning the costs of actions such that…

Authors: Yihao Hou、Christoph Bert、Ahmed Gomaa、Godehard Lahmer、Daniel Hoefler、Thomas Weissmann、Raphaela Voigt、Philipp Schubert、Charlotte Schmitter、Alina Depardon、Sabine Semrau、Andreas Maier、Rainer Fietkau、Yixing Huang、Florian Putz Paper: https://arxiv.org/abs/2408.10715 Introduction In the rapidly evolving field of artificial intelligence (AI), large language models (LLMs) have shown significant promise across various domains, including medicine. However, the integration of these models into clinical practice, particularly for tasks such as generating physician letters, presents unique challenges. This study, conducted by Yihao Hou and colleagues, investigates the fine-tuning of LLaMA-3, a large language model, for generating physician letters in radiation oncology while preserving patient privacy. The primary goal is to enhance efficiency in clinical documentation…

Authors: Sagar Uprety、Amit Kumar Jaiswal、Haiming Liu、Dawei Song Paper: https://arxiv.org/abs/2408.10711 Introduction Large Language Models (LLMs) such as GPT-4 and the LLaMa family have significantly advanced the field of Artificial Intelligence (AI) by enhancing the ability to comprehend and generate natural language text. These models are increasingly being deployed in real-world scenarios, making decisions and taking actions based on their understanding of the context. However, ensuring that these decisions align with human values and user expectations is a complex challenge, as human values and decisions are often influenced by cognitive biases. This study investigates the alignment of LLMs with human judgments, specifically…

Authors: Andrew Kiruluta、Andreas Lemos Paper: https://arxiv.org/abs/2408.09967 Introduction Background Linear programming (LP) has been a fundamental optimization technique since its inception by Dantzig in 1947. It is widely used in various fields such as operations research, economics, and engineering due to its ability to optimize objectives subject to linear constraints. However, traditional LP methods face limitations when dealing with non-linear, high-dimensional, and dynamic environments. On the other hand, machine learning (ML), particularly deep learning, has shown remarkable success in modeling complex patterns and making predictions from large datasets. Despite these strengths, ML models often lack the explicit interpretability and rigorous constraint…

Authors: Yuxiong Xu、Jiafeng Zhong、Sengui Zheng、Zefeng Liu、Bin Li Paper: https://arxiv.org/abs/2408.09933 Introduction Background The rapid advancements in Artificial Intelligence Generated Content (AIGC) have significantly improved the naturalness, fidelity, and variety of synthetic speech. However, these advancements have also led to an increase in sophisticated forgeries that are nearly indistinguishable from authentic speech to the human ear. This poses a significant threat to Automatic Speaker Verification (ASV) systems, which are increasingly vulnerable to spoofing and deepfake attacks. These attacks can convincingly simulate a target speaker’s voice, presenting substantial societal risks. Problem Statement The ASVspoof challenges have been instrumental in driving the development of…

Authors: Chunyang Jiang、Chi-min Chan、Wei Xue、Qifeng Liu、Yike Guo Paper: https://arxiv.org/abs/2408.09849 Introduction Large Language Models (LLMs) have demonstrated exceptional capabilities across a wide range of natural language processing (NLP) tasks. However, fine-tuning these models using high-quality datasets under external supervision is often prohibitively expensive. To address this, self-improvement approaches for LLMs have been developed, where the models are trained on self-generated data. Despite the potential of this approach, the variability in data quality poses a significant challenge. This paper introduces a novel metric called Distribution Shift (DS) weight, inspired by Importance Weighting methods, to filter out self-generated samples with high distribution shift…

Authors: Zhirong Huang、Shichao Zhang、Debo Cheng、Jiuyong Li、Lin Liu、Guixian Zhang Paper: https://arxiv.org/abs/2408.09646 Introduction Recommender systems have become an integral part of many online services, from e-commerce platforms like Amazon to streaming services like Netflix. These systems aim to predict user preferences and recommend items that align with those preferences. However, traditional recommender systems often suffer from biases, particularly popularity bias and conformity bias. Popularity bias leads to the over-representation of popular items, while conformity bias causes users to align their choices with the group, even if it conflicts with their personal preferences. This paper introduces a novel framework, Debiased Contrastive Learning for…